CN1278501C - Compressing method and compressing equipment for mobile target-position data - Google Patents
Compressing method and compressing equipment for mobile target-position data Download PDFInfo
- Publication number
- CN1278501C CN1278501C CN02154031.4A CN02154031A CN1278501C CN 1278501 C CN1278501 C CN 1278501C CN 02154031 A CN02154031 A CN 02154031A CN 1278501 C CN1278501 C CN 1278501C
- Authority
- CN
- China
- Prior art keywords
- target
- error
- linear
- moving target
- current location
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99941—Database schema or data structure
- Y10S707/99942—Manipulating data structure, e.g. compression, compaction, compilation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The present invention discloses a method and equipment for compressing the position data of a plurality of mobile targets in continuous time. The method comprises the steps that a) the current position data of the mobile targets is received; b) whether the targets are in a linear movement state is distinguished; c) if that targets do linear movement in the past is judged in the b), whether the targets still do the linear movement after the current position is added is judged; if the targets still do the linear movement, the positions of the targets are not updated; d) if that the linear movement is destroyed after the current position is added is judged in the c), the positions of the targets are updated, and go to the a); e) if that the targets do not do the linear movement in the past is judged in the b), whether the targets randomly move nearby is distinguished; if truth, the positions of the targets are not updated; else, the positions of the targets are updated, the targets are set to be in the linear movement state, and go to the a), wherein in the position update of the targets, a position queue with variable length is selected, and the compressed target positions are stored to the position queue; then, the position queue inserts the positions of the targets in different time to a database at inserting speed which can be received by the database.
Description
Technical field
The present invention relates to the compression method and the compression device of moving target position data, more particularly, relate under the situation that guarantees the inquiry precision, compress the method and the equipment thereof of position data of the continuous time of a large amount of moving targets.
Background technology
Location for moving target has proposed several different methods.For example, global positioning system (GPS), its positioning accuracy is very high, and (the gps satellite constellation is launched two kinds of spreading codes earthward: C/A sign indicating number and P sign indicating number.The C/A sign indicating number is civilian sign indicating number, and positioning accuracy is low, about 100 meters, adopts the differential technique precision can bring up to 10 meters.High-precision P sign indicating number is military sign indicating number, and its positioning accuracy is less than 10 meters, and rate accuracy is 0.06-0.1m/s).In addition, present mobile phone also can be located by the base station, and (positioning accuracy depends on the radius of sub-district, place, as in areas of Beijing but positioning accuracy is lower, base station density is higher, and cell-of-origin (Cell of Origin) technological orientation precision can reach about 200 meters; And in the suburb, base station density is lower, and this location technology positioning accuracy can only reach one or two kilometer).
Present existing system utilizes these location technologies that moving target is located continuously, and the positional information that obtains put into database, thereby, can carry out the inquiry of variety of way to the position of moving target in the database, we are called moving target database (Moving ObjectDatabase) with whole such system.Therefore, as shown in Figure 1, these moving target Database Systems comprise from hardware: the server at mobile unit (Mobile Unit), location facilities, communication interface and database place.Mobile unit be one be installed in can be monitored on the moving target equipment, as mobile phone, GPS receiver etc.When using the GPS location, location facilities and mobile unit usually are referred to as mobile unit.They send the position of moving target into database by communication interface, and database provides external inquiry and other service interface.
Below provide some data base querying examples of applications: when a people entered scope from 500 meters of McDonald, service provider gave his electronic coupons of mobile phone; The taxi driver handles the position of inquiry all taxi yesterday morning 9:00 of its company to 2:00 in afternoon this day; Or the like.The moving target database can be used for taxi management, logistics, location-based service or the like.
In such system, if database storage adopts the discrete point model, it is the position of data-base recording moving target discrete time point, want the position of a large amount of moving target of real-time tracking, perhaps the frequency of tracking target is very high, database will insert the position of a large amount of moving targets in the unit interval, database holds on the one hand can't stand high like this insertion speed, can take a large amount of memory spaces on the other hand.Here we are example with taxi management moving target Database Systems, explain this problem: a taxi company has 30,000 taxi, per 1 minute to a taxi location once, the position sends to database, the database per second will carry out 495 positions and insert operation, and database can't support high like this data to insert operation, and such moving target Database Systems just can not normal operation.
The another one problem is a memory space problem, generally uses technology such as linear interpolation or match at the position enquiring of non-sampling instant, is example with the linear interpolation, and target is at T (T
1≤ T≤T
2) constantly position X is that this target is at T
1Moment position X
1And T
2Moment position X
2Linear interpolation, be shown below.
Fig. 2 has shown the error of position enquiring, is the distance of this point interpolation position and this actual position.
Consider and in database, use two-dimensional position that record is carried out in the taxi position of this taxi company, each position takies 16 bytes, the data volume that provides the company of this service will store 659.1M every day for this taxi company so, be the 19.3G data volume every month, the annual data volume of wanting 235G, this has still only considered the situation of clear data storage, does not also consider the data base administration aspect, as index etc.So big data storage capacity is also relatively more difficult to the data library management, and the expensive of memory space also can not be ignored.
If we can compress such data, the data after the compression are put into database, just can address the above problem effectively.Compression can be considered from the following aspects:
(1) as U.S. Pat 6,327, that uses among the 533B1 removes redundant filter method, set the threshold value of time or displacement, if the time of this target travel or displacement do not surpass this threshold value, just not to the position record of moving target, can reach the purpose of compression like this, and be recorded in the position that remains target in the database, query processing is very convenient, but when a large amount of moving targets all in when motion, this method lost efficacy, the insertion operation of database is still a lot, and system can not normally move probably;
(2) the constant speed linear movement is a kind of very important motion mode of moving target, in U.S. Pat 5, also mentioning linear movement in 187,689 can be used this linear prediction method also can reduce memory space by linear prediction, and be recorded in the position that remains target in the database, query processing is very convenient, but when a large amount of moving targets all during migration nearby, compression ratio is very low, the insertion operation of database is still a lot, and system can normally not move probably yet;
(3) movement locus of target is regarded as a kind of curve, these curves are carried out curve fitting or conversion, the coefficient of institute's match, or the coefficient storage after the conversion is in database, also can reach the purpose of compression, but this mode has following two shortcomings: (a) relation between the mathematics form of curve after match and the conversion and the sum of errors curve between the virgin curve is normally very complicated, be difficult to obtain, and the curve fit of use fixed form, precision was difficult to guarantee when the position was inquired about; (b) also to do complicated inverse transformation during inquiring position, influence the time of query processing greatly.The query manipulation of a lot of databases all can not directly use.
Summary of the invention
In order to address the above problem, the invention provides a kind of method and apparatus that the continuous position data of a large amount of moving targets are compressed, its receives the position of the continuous motion of a large amount of moving target that transmits from mobile unit or positioning equipment, these position entrance pressures are contracted, positional information after the compression is inserted in the database, solve database and insert the problem that to move under the frequency, and guaranteeing to reduce the memory space of continuous position under certain position enquiring precision at height.
According to an aspect of the present invention, provide a kind of method of compressing a large amount of moving targets in the position data of continuous time, comprise step: the current location data that a) receives moving target; B) differentiate this target and whether be in the linear movement state in the past; C) if judge that in step b) target moves linearly in the past, whether judgement still moves linearly after adding current location, if still move linearly, does not upgrade its position; D) if linear movement is destroyed judge the adding current location in step c) after, then upgrade its position, return step a); And e) if judging target in step b) does not move linearly in the past, differentiate its whether nearby random walk, if, do not upgrade its position, otherwise upgrade its position, and set this target and be in the linear movement state, return step a).
The a large amount of moving targets of compression according to the present invention are in the method for the position data of continuous time, its position of described renewal is meant, select an adjustable length position queue, store the moving target position that will insert in the database after the compression into this variable position formation, described then position queue is inserted speed with the receivable data of database, inserts the different positions constantly of moving target in database.
According to a further aspect in the invention, provide a kind of equipment that compresses the moving target position data, comprising: input interface is used to receive the current location of the moving target of continuous motion; Data compression device is used for the positional information compression by input interface received; And output interface, be used to export the moving target position after the compression.Wherein, described data compression device comprises: the linear model processing unit is used for when moving target is in the linear movement state in the past, whether still move linearly after judging the adding current location, if still move linearly, do not upgrade its position, otherwise upgrade its position; And simple threshold values models treated unit, be used for when moving target does not move linearly, differentiate its whether nearby random walk, if do not upgrade its position, otherwise upgrade its position.
In the method and equipment thereof that the continuous position data of a large amount of moving targets are compressed according to the present invention, when receiving the positional information of a target, differentiate this target earlier and whether be in the past the linear movement state, whether still move linearly after adding this position, if still move linearly, in position queue, do not upgrade its position; If linear movement is destroyed, in position queue, upgrade its position; If target is not moving linearly in the past, differentiate its whether nearby random walk, if, in position queue, not upgrade its position, otherwise in position queue, upgrade its position, whole process can automatically switch.Position queue is inserted speed with the receivable data of database then, inserts the different positions constantly of moving target in database.Thereby, can improve the compression ratio of data, reduce the insertion operation of database, guarantee the normal operation of system.Simultaneously, remove complicated operations such as conversion such as match and inverse transformation thereof from, can reduce the time of query processing.
Description of drawings
By following detailed description by accompanying drawing, will more easily understand the present invention, wherein identical label is specified the unit of same structure, and therein:
Fig. 1 illustrates the structure of moving target Database Systems;
Fig. 2 illustrates the schematic diagram of position enquiring error;
Fig. 3 illustrates the principle that position queue is handled;
Fig. 4 illustrates the flow chart of the position data after the compression being put into the processing method of position queue;
Fig. 5 illustrates the flow chart that the position data in the position queue shown in Figure 3 is inserted the method for database;
Fig. 6 is illustrated under the situation of road topology, the continuous position of the moving target of acquisition;
Fig. 7 illustrates the moving target position through overcorrect is carried out the schematic diagram that linear prediction and predicated error exceed the threshold value situation;
Fig. 8 is illustrated under the situation that does not have road topology, the continuous position of the moving target of acquisition and the judgement of linear movement;
Fig. 9 illustrates moving target speed when changing suddenly, the flow chart of the processing method that the position data of moving target is compressed;
Figure 10 illustrates the situation of simple threshold values model;
Figure 11 is illustrated in the off-limits situation of error under the simple threshold values model condition;
When Figure 12 illustrates moving target and fluctuates near a certain position, the flow chart of the processing method that the position data of moving target is compressed;
Figure 13 illustrates the general flow chart according to the processing method that the continuous position data of a large amount of moving targets are compressed of the present invention;
Figure 14 illustrates the block diagram according to the equipment that the continuous position data of a large amount of moving targets are compressed of the present invention;
Figure 15 is illustrated under the situation of the topological sum physical structure that stores road in the memory cell, the block diagram of linear model processing unit; And
Figure 16 is illustrated under the situation of the topological sum physical structure of not storing road in the memory cell, the block diagram of linear model processing unit.
Embodiment
Describe the preferred embodiments of the present invention in detail below in conjunction with accompanying drawing.
In the method and equipment thereof that the continuous position data of a large amount of moving targets are compressed according to the present invention, the position of moving target can be from plurality of devices, as the GPS receiver, wireless network or the like, and a target can be from different equipment in difference position constantly, but all places has been projected into the plan position approach, and the positional information of acquisition can have different precision.
In the method and equipment thereof that the continuous position data of a large amount of moving targets are compressed according to the present invention, can store the road topology and the physical structure in a large amount of target travels zone in advance, perhaps do not store this background knowledge.
In the method and equipment thereof that the continuous position data of a large amount of moving targets are compressed according to the present invention, at first the continuous position to moving target compresses, and insert in the database position after will compressing then.What store in the database is the position of each moving target in each discrete moment point.
Describe in detail according to the method that the continuous position data of a large amount of moving targets are compressed of the present invention below in conjunction with accompanying drawing.This method can be divided into the two large divisions on the whole, i.e. the processing of position formation and the compression of position data.
1. the processing of position queue
Fig. 3 illustrates the principle that position queue is handled.In position queue processing method according to the present invention, an adjustable length position queue is set, be used for the position of the moving target that will insert database after the store compressed, it can be along with increasing in certain limit of the position that receives is elongated.When certain target position constantly has been inserted in the database among the position queue, then from position queue, leave out this position.Position queue can be stored in the internal memory, also can be stored in one or more files.
Fig. 4 illustrates the flow chart of the position data after the compression being put into the processing method of position queue.At first, at step SP41, receive the position data after the compression.At step SP42, judge whether the position of formation enough can join this position data that receives at step SP41 in the position queue greatly then.If the result who judges in step SP42 is a "Yes", then handle and advance to step SP44, the position data that is received is put into position queue.If the result who judges is a "No", then in step SP43,, and in step SP44, the data that received are put into position queue with the position queue lengthening in step SP42.Afterwards, handle advancing to step SP45, wait for receiving next position data.
Fig. 5 illustrates the flow chart that the position data in the position queue shown in Figure 3 is inserted the method for database.As shown in Figure 5, at first in step SP51, judge whether position queue is empty, if there are not data in the position queue, i.e. position formation is empty, then waits at step SP52, until till the position queue non-NULL.Then, handle to advance to step SP53, the position data of last position in the position queue is inserted in the database.Afterwards, handle to turn back to step SP51, continue to judge whether that new position data adds in the position queue.
The continuous position data of more than having described after how will compressing are inserted in the database.But as described in front, before described continuous position data are inserted database, the continuous position data compression could must be reduced the insertion operation of database, guarantee the normal operation of system.To describe the compression method of continuous position data of the present invention below in detail.
2. the compression of position data
There is certain error in the location of moving target, is having under the situation of road usually, and the continuous position of the moving target of acquisition as shown in Figure 6.The position location of the moving target received is represented in filled circles position among Fig. 6, and the open circles position is the actual position of target.
Be described at first in two kinds of situation below and how judge linear movement in the linear prediction model.
(a) if stored the information such as topological sum physical structure, position of road, and road graticule dense degree little (two the road minimum distances in space are greater than the position error of twice), then can utilize spatial index to find the correction position of moving target.For the position that GPS obtains, also can use bearing calibration, it is corrected to road gets on, thereby obtain the correction position of moving target on road, shown in the open circles position among Fig. 6.
Fig. 7 illustrates the schematic diagram that carries out linear prediction to through the moving target position of above-mentioned correction, and wherein filled circles is represented the position behind the moving target sampling correcting.
If it is smaller that target speed changes, be predicted as example with 2 first-order linears, utilize T
1Moment position X
1And T
2Moment position X
2Can dope target at T
3Position X constantly
3', that is:
X
3′＝2X
2-X
1
So just obtained predicated error
Err(T
3)＝|X
3′-X
3|
According to the error of both direction, just can judge the magnitude relationship of total predicated error and linear prediction threshold value.Can certainly use the multiple spot prediction, promptly
X
n＝f(X
n-1，X
n-2，...，X
1)
But it is better that our experiment shows with the first-order linear prediction effect.
Among the figure, when to T
5Position X constantly
5', Forecasting Methodology below using:
X
5′＝2X
4′-X
3′
And do not use X
4And X
3Prediction X
5, in last figure, the position of inserting in the database comprises X like this
1, X
2And X
5Thereby, insensitive to the fluctuation of velocity magnitude, the compression ratio height.
(b) under the situation that does not have information such as the topological sum physical structure of road, position, perhaps these information is being arranged, but the close integrated level of road graticule does not satisfy above-mentioned condition, under the not high again situation of positioning accuracy, if according to the position data of the moving target of being received, we obtain moving target X
1And X
2The position, but the error of these positions is may local distribution inhomogeneous, for example X
1The position possible error is very little, X
2Site error is but very big, causes velocity attitude to estimate very big deviation, and is less even the target velocity absolute value changes, and moves on road, and the target of prediction is at T
3Position X constantly
3Deviation also very big, just do not reach compression effects.
At this moment, we open up an interim memory space to each target, the position that linear prediction is satisfied in storage.As shown in Figure 8, be example with the directions X, at first we have obtained X
1And X
2Interim memory space is put in the position, and 2 can unique definite straight line, a bit satisfies the linear prediction sign before therefore putting, and works as X
3When arrive in the position, if each point satisfied linear prediction in the past, just to X
1, X
2, X
3Carry out fitting a straight line, obtain the expression of straight line L1:
X(t)＝K
1·t+B
1
T represents the time in the formula, can be with X
1, X
2, X
3Time corresponding substitution following formula in position just obtains the error of the fitting a straight line of each point correspondence.Can obtain the error of Y direction with example, thereby obtain the composition error of each point, if the each point composition error all in the threshold range of regulation, X then
1, X
2, X
3Satisfy linear prediction, X
3Point needn't deposit position queue in.X in like manner
4That puts also needn't store position queue into to straight line L2.
But, work as X
5After putting adding, the composition error of some point is understood in the threshold range of overshoot, then X
5Point has destroyed linear prediction, will be stored in position queue, at this moment needs to empty the interim memory space of position of the satisfied linearity of this target.
Below discuss how to judge that target is in linear movement in two kinds of situation, but linear movement also should add following processing.If moving target speed has bigger variation suddenly, store as described above in the linear determination methods of topological sum physical structure of road, at T
1-T
5The position of inquiry moving target can be on the dotted line of Fig. 7 constantly, and obviously, error is very big, if record previous moment position X
4, the precision of position enquiring is just very high.But if always later write down the position of previous moment, compression ratio can reduce again greatly.
For addressing this problem, in the method that the continuous position data of a large amount of moving targets are compressed according to the present invention, store following data at each moving target:
The current location of real-time this target that receives, real-time update is called PresentLoc;
This target last record position in database, real-time update is called LastLocInDB;
The correction position of the previous moment of moving target sampling current location, real-time update is called LastLocSamp, promptly arrives X among Fig. 7
5The LastLocSamp=X during position
4', when adopting linear fit, also be the position of match point;
When storing information such as the topological sum physical structure of road, position, the correction position of LastLocSamp previous moment, real-time update is called BeforeLastLocSamp;
Moving target satisfies the sampling number of linear prediction, and real-time update is called PredAddNum;
Moving target linear prediction error threshold value according to the position enquiring precision set, is called PredThresh, is definite value;
The special threshold value of inquiry precision set another one as requested is called PredSpecilThresh, is definite value; And
The sampling number threshold value according to the inquiry precision set, is called NumThresh, is definite value.
If the site error of linear prediction is greater than threshold value, but some moment of front satisfy linear prediction model, as shown in Figure 7, if only write down T
1, T
2, T
5Position constantly is then at moment T
4During inquiring position, error will go beyond the scope.At this moment, adopt the method shown in the flow chart of Fig. 9 to come the packed data of needs storage is done further processing.The method is in order further to improve the position enquiring precision, if the position enquiring precision of application requirements is not very high, can to ignore this step.
As shown in Figure 9, at first judge that at step SP91 whether predicated error is greater than special threshold value.If predicated error is not more than the special threshold value that sets, then handle advancing to step SP92, judge and satisfy the counting of linear prediction whether greater than the sampling number threshold value.If satisfy counting of linear prediction greater than the sampling number threshold value, then handle and advance to step SP94, in database, insert LastLocSamp and PresentLoc.If judge that in step SP92 satisfying counting of linear prediction is not more than the sampling number threshold value, then handle advancing to step SP95, PresentLoc is inserted in the database.
If judge that in step SP91 predicated error is greater than setting special threshold value, then handle and advance to step SP93, judge whether LastLocInDB equals LastLocSamp, if LastLocInDB and LastLocSamp are unequal, then in step SP94, LastLocSamp and PresentLoc are inserted in the database.If judge that in step SP93 LastLocInDB equals LastLocSamp, then handle directly advancing to step SP95, PresentLoc is inserted in the database.
Then, after handling step SP94 or SP95, advance to step SP96, at this, clear linear movement sign, with satisfying the zero clearing of counting of linear movement, empty BeforeLastLoc, make center Center, LastLocInDB and the LastLocSamp of near the target location of migration a certain position equal PresentLoc.And only keep PresentLoc at the interim memory space that satisfies the linear prediction position.
When the above moves linearly for moving target, the compression that its position data is done is handled.In the following description processing shown in Figure 9 is called and handles 1.
To describe below when moving target is paced up and down near a certain position, the compression that its position data is carried out will be handled.Figure 10 is illustrated in the condition that near target location of fluctuating, a certain position is not upgraded, and in order to narrate conveniently, we are called the simple threshold values model.
In the simple threshold values model, if the error between the center of the position of target and its regulation less than preset threshold, is not upgraded its position, as the target location among Figure 10 1 in position queue; Otherwise, in position queue, upgrade its position, as the target location among Figure 10 2.
For certain target, we need set its simple threshold values model center position, and real-time update is called Center; And set its simple threshold values model threshold, according to the position enquiring precision set, definite value is called SampThresh.
Also there be the problem same with linear prediction model in the simple threshold values model.That is, satisfy the simple threshold values model, as shown in figure 11, suppose that to T8 constantly the time, threshold model is destroyed,, will turn back on the position of straight line L during the position constantly in the middle of the inquiry if only write down the T1 and the position in two moment of T8 this moment when longer a period of time.Obviously, the inquiry precision can go beyond the scope.
At this,, the processing employing and the linear prediction model of simple threshold values model are handled same principle according to the method that the continuous position data of a large amount of moving targets are compressed of the present invention.Also set special threshold value and satisfy counting of simple threshold values model, and accumulative total satisfies counting of simple threshold values model.
When Figure 12 illustrates moving target and fluctuates near a certain position, the flow chart of the processing method of when target location exceeds threshold range the position data of moving target being compressed.
As shown in figure 12, judge in step SP1201 whether LastLocInDB equals LastLocSamp, do not satisfy the point of simple threshold values model promptly, if LastLocInDB is not equal to LastLocSamp, then, handle then and advance to step SP1205 at the mid-linear movement sign of step SP1203.
If judge that in step SP1201 LastLocInDB equals LastLocSamp, then handle and advance to step SP1202, whether judge site error with Center greater than special threshold value,, then handle advancing to step SP1205 if be not more than special threshold value with the site error of Center.If the site error of judgement and Center is greater than special threshold value in step SP1202, then handle and advance to step SP1204, judge and to satisfy the counting of simple threshold values model whether greater than the threshold value of counting, if satisfy counting of simple threshold values model greater than the threshold value of counting, then handle and advance to step SP1206, be not more than the threshold value of counting if satisfy counting of simple threshold values model, then handle advancing to step SP1205.
In step SP1205, PresentLoc is inserted in the database.In step SP1206, earlier LastLocSamp is inserted in the database, again PresentLoc is inserted in the database, and put and satisfy the linear movement sign.
After the processing of completing steps SP1205 or SP1206, advance to step SP1207, will satisfy the zero clearing of counting of simple threshold values model, and carry out last position data BeforeLastLoc and handle, it is emptied; Make Center=PresentLoc, LastLocInDB=PresentLoc, LastLocSamp=PresentLoc; And, in satisfying the interim memory space of linear prediction position, keep LastLocSamp and PresentLoc according to the linear movement sign.
Here, the position enquiring precision that satisfies the special threshold value of simple threshold values model and requirement is relevant, can be identical with the special threshold value of the linear prediction of front, and the threshold value of counting also can be identical, also can set in addition.When in the following description top moving target illustrated in fig. 12 being fluctuateed near a certain position, the processing method that the position data of moving target is compressed is called processing 2.
Description above comprehensive provides the automatic switchover of how carrying out between simple threshold values model and two models of linear prediction below in conjunction with flow chart shown in Figure 13, thereby obtains the target position data of high compression ratio, and impact position is not inquired about precision.
Figure 13 illustrates the general flow chart according to the processing method that the continuous position data of a large amount of moving targets are compressed of the present invention.
As shown in figure 13, carry out initialization, necessary threshold parameter is set at step SP1301.Then, in step SP1302, from location facilities or mobile unit, receive the PresentLoc of a certain moving target.Judge in step SP1303 then whether this target data is the position data of coming for the first time, if next for the first time position data, then handle and advance to step SP1304, make Center=PresentLoc, LastLocInDB=PresentLoc, LastSamp=PresentLoc, afterwards, return step SP1302, continue from location facilities or mobile unit receiving position data.
If judge that at step SP1303 the target data that is received is not the position data of coming for the first time, then handle advancing to step SP1305, judge whether last position satisfies linear prediction model.If linear prediction model is not satisfied in last position, judge in step SP1307 that then whether the simple threshold values model error is greater than the threshold value that sets.
, then handle and advance to step SP1312 greater than the threshold value that sets if in step SP1307, judge the simple threshold values model error, carry out aforesaid processing 2.Otherwise, handle advancing to step SP1313, simple threshold values model counting is added 1.
After the processing of the processing 2 of execution of step SP1312 or step SP1313, return step SP1302, continue from location facilities or mobile unit receiving position data.
Satisfy linear prediction model if in step SP1305, judge last position, then handle advancing to step SP1306, judge whether current location also satisfies linear prediction model after adding.If also satisfy linear prediction model after current location adds, then handle and advance to step SP1308, make Center=PresentLoc, upgrade LastLocSamp, BeforeLastLocSamp and interim memory space.Afterwards, return step SP1302, continue from location facilities or mobile unit receiving position data.
If in step SP1306, judge, after adding, current location no longer satisfies linear prediction model, then handle advancing to step SP1309, judge that whether the simple threshold values model error is greater than the threshold value that sets.If the simple threshold values model error is not more than the threshold value that sets, handle and advance to step SP1310, carry out following processing 3: LastLocSamp is inserted in the database; Clear linear movement sign; Simple threshold values model counting adds 1; LastLocSamp upgrades; BeforeLastLocSamp upgrades; And interim memory space upgrades.
, then handle and advance to step SP1311 greater than the threshold value that sets if in step SP1309, judge the simple threshold values model error, carry out above-mentioned processing 1.
After the processing 1 of the processing 3 of execution of step SP1310 or step SP1311, return step SP 1302, continue from location facilities or mobile unit receiving position data.
Here the setting of all threshold values all is comprehensive afterwards decision of factors such as compression ratio of considering the position enquiring precision of positioning accuracy, application, requirement and need reaching.Do location-based advertisement for common mobile phone location, set 200 meters-300 meters threshold value and just can reach compression ratio more than 5 times, promptly can reach the requirement of general application.
Below describe the method that the continuous position data of a large amount of moving targets are compressed of the present invention in conjunction with the accompanying drawings, will describe the equipment that the continuous position data to a large amount of moving targets that realize said method are compressed below.
As shown in figure 14, the equipment that the continuous position data of a large amount of moving targets are compressed according to the present invention comprises: input interface, be used to receive the position of the continuous motion of a large amount of moving target of sending here from location facilities or mobile unit, these positional informations comprise moving target, positioning accuracy, current location etc.; Data compression device is used for the positional information compression by input interface received; And output interface, the moving target position after being used for compressing sends to database or records file.
By in described data compression device, setting different threshold values, the precision that can guarantee customer requirements is inquired about in the position after the compression.
Described data compression device comprises: memory cell, simple threshold values models treated unit and linear model processing unit.
Memory cell comprises: the position queue memory is used for will being inserted into after the store compressed position of database; Threshold value memory is used to store the threshold value of simple threshold values model, the prediction error threshold of linear prediction model and special threshold value, the position number accumulative total threshold value of linear prediction model; The intermediate object program temporary storage is used to store PresentLoc, Center, LastLocInDB, LastLocSamp, the PredAddNum of each target and satisfies linear prediction model sign etc. with the front position.Differentiate the temporary storage that linear method also will comprise the position of the satisfied linearity of each target for using linear fit.
The linear model processing unit is used for when moving target is in the linear movement state in the past, and whether judgement still moves linearly after adding current location, if still move linearly, does not upgrade its position, otherwise upgrades its position.
Figure 15 is illustrated under the situation of the topological sum physical structure that stores road in the memory cell, the block diagram of linear model processing unit.As shown in figure 15, in this case, this linear model processing unit comprises: the linear prediction parts are used to utilize the position in former some moment of target that linear prediction is carried out in the position of target current time; The predicated error calculating unit is used to calculate predicated error; And judging part, if predicated error in the linear prediction threshold range, confirms then that moving target still moves linearly.
Do not store in memory cell under the situation of topological sum physical structure of road, the structure of described linear model processing unit as shown in figure 16.In this case, this linear model processing unit comprises: the match parts, and match is carried out in the position that is used to utilize current location and target to be in linear movement in the past, obtains the position-time linear equation of match; The error of fitting estimation section, the time that is used for current time and target are in linear movement in the past is the position-time linear equation of substitution institute match respectively, estimates the error of fitting of each time point; And judging part, if the error of fitting of all time points all within linear range of predicted value, confirms that then moving target still moves linearly.
Whether nearby simple threshold values models treated unit is used to carry out above-mentioned processing 2, promptly when moving target does not move linearly, differentiate its random walk, if do not upgrade its position, otherwise upgrade its position.Simple threshold values models treated unit comprises simple threshold values model comparator and simple threshold values modal position Error Calculator.
Below described the preferred embodiments of the present invention in conjunction with the accompanying drawings, but the present invention is not limited in this concrete execution mode.Under the situation of the spirit and scope that do not depart from claim, can make various changes to it.
Claims (17)
1. a large amount of moving targets of compression comprise step in the method for the position data of continuous time:
A) current location data of reception moving target;
B) differentiate this target and whether be in the linear movement state in the past;
C) if judge that in step b) target moves linearly in the past, whether judgement still moves linearly after adding current location, if still move linearly, does not upgrade its position;
D) if linear movement is destroyed judge the adding current location in step c) after, then upgrade its position, return step a); And
E) if judging target in step b) does not move linearly in the past, differentiate its whether nearby random walk, if, do not upgrade its position, otherwise upgrade its position, and set this target and be in the linear movement state, return step a).
2. compression method as claimed in claim 1, it is characterized in that: its position of described renewal is meant, select an adjustable length position queue, store the moving target position that will insert in the database after the compression into this variable position formation, described then position queue is inserted speed with the receivable data of database, inserts the different positions constantly of moving target in database.
3. compression method as claimed in claim 2 wherein, is stored topological sum physical structure, the positional information of road in advance, and this method further comprises:
During greater than the position error of twice, before step b), utilize spatial index to obtain the correction position information of moving target at two road, space minimum distance.
4. compression method as claimed in claim 3 is characterized in that, and is further comprising the steps of in step c):
C1) utilize the position in former some moment of target that linear prediction is carried out in the position of target current time;
C2) calculate predicated error; And
C3) if predicated error in linear prediction error threshold scope, confirms that then moving target still moves linearly.
5. compression method as claimed in claim 4 is characterized in that, described linear prediction is 2 first-order linear predictions or the prediction of multiple spot first-order linear.
6. compression method as claimed in claim 2 is characterized in that, and is under the situation of the topological sum physical structure of not storing road in advance, positional information, further comprising the steps of in step c):
C11) match is carried out in the position that utilizes current location and target to be in linear movement in the past, obtains the position-time linear equation of match;
C12) time that current time and target is in the in the past linear movement position-time linear equation of substitution institute match is respectively estimated the error of fitting of each time point; And
C13) if the error of fitting of all time points all within linear prediction error threshold values scope, confirm that then moving target still moves linearly, and this target be made as its correction position in the match position of each time point.
7. as claim 5 or 6 described compression methods, it is characterized in that: in step d), comprise the following steps:
D1) if linear movement is destroyed judge the adding current location in step c) after, then judge whether nearby random walk of target, if, upgrade its position, set this target and be near random walk state;
D2) if in steps d 1) in judge that target is not near a random walk state being in, judge that then predicated error or error of fitting are whether greater than according to the special error threshold of the target of position enquiring precision set, if predicated error or error of fitting are greater than according to the special error threshold of the target of position enquiring precision set, judge whether last record position is identical with the correction position of last position, if it is identical, then write down current location, otherwise, write down the correction position and the current location of last position;
D3) if predicated error or error of fitting are not more than according to the special error threshold of the target of position enquiring precision set, then judge and satisfy the counting of linear prediction whether greater than the threshold value of counting, if be not more than the threshold value of counting, then write down current location, otherwise, write down the correction position and the current location of last position; And
D4) only keep current location at the interim memory space that satisfies the linear prediction position.
8. as the arbitrary described compression method of claim 1 to 5, it is characterized in that, in step e), comprise the following steps:
E1) judge whether last record position equals the correction position of last position,, then put the linear movement sign and write down current location if unequal;
E2) if at step e1) in judge that last record position equals the correction position of last position, judge that then whether site error is greater than the special threshold value according to the position enquiring precision set, if site error is not more than the special threshold value according to the position enquiring precision set, then only write down current location;
E3) if at step e2) in judge that site error is greater than the special threshold value according to the position enquiring precision set, then judge and satisfy the counting of simple threshold values model whether greater than the threshold value of counting, if satisfy counting of simple threshold values model greater than the threshold value of counting, then write down the correction position and the current location of last position, otherwise, only write down current location; And
E4) will satisfy the zero clearing of counting of simple threshold values model, empty last position data,, in satisfying the interim memory space of linear prediction position, keep the correction position and the current location of last position according to the linear movement sign.
9. as the arbitrary described compression method of claim 2 to 6, it is characterized in that described position queue is opened up in internal memory, perhaps the form with one or more files exists.
10. as the arbitrary described compression method of claim 1 to 6, it is characterized in that, before step a), also comprise initialization step, receive and write down preceding several position datas of moving target, judge that according to the preceding several position datas that received targets are in the state that the linear movement state still is near random walk then.
11. an equipment that compresses the moving target position data comprises:
Input interface is used to receive the current location of the moving target of continuous motion;
Data compression device is used for the positional information compression by input interface received; And
Output interface is used to export the moving target position after the compression;
Wherein, described data compression device comprises:
The linear model processing unit is used for when moving target is in the linear movement state in the past, and whether judgement still moves linearly after adding current location, if still move linearly, does not upgrade its position, otherwise upgrades its position; And
Whether nearby simple threshold values models treated unit is used for when moving target does not move linearly, differentiate its random walk, if do not upgrade its position, otherwise upgrade its position.
12. compression device as claimed in claim 11 also comprises storage device, is used to store topological sum physical structure, the positional information of road and is the necessary intermediate object program of data compression.
13. compression device as claimed in claim 12 is characterized in that, described linear model processing unit comprises:
The linear prediction parts are used to utilize the position in former some moment of target that linear prediction is carried out in the position of target current time;
The predicated error calculating unit is used to calculate predicated error; And
Judging part is if predicated error in linear prediction error threshold scope, confirms that then moving target still moves linearly.
14. compression device as claimed in claim 13 is characterized in that, described linear prediction is 2 first-order linear predictions or the prediction of multiple spot first-order linear.
15. compression device as claimed in claim 11 is characterized in that, described linear model processing unit comprises:
The match parts, match is carried out in the position that is used to utilize current location and target to be in linear movement in the past, obtains the position-time linear equation of match;
The error of fitting estimation section, the time that is used for current time and target are in linear movement in the past is the position-time linear equation of substitution institute match respectively, estimates the error of fitting of each time point; And
Judging part is if the error of fitting of all time points all within linear prediction error threshold scope, confirms that then moving target still moves linearly.
16. as the arbitrary described compression device of claim 11 to 15, it is characterized in that, described output interface is an adjustable length position queue, it is being stored and will insert moving target position in the database after the compression, and, in database, insert the different positions constantly of moving target with the receivable data insertion of database speed.
17. compression device as claimed in claim 16 is characterized in that, described position queue is opened up in internal memory, and perhaps the form with one or more files exists.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN02154031.4A CN1278501C (en) | 2002-12-06 | 2002-12-06 | Compressing method and compressing equipment for mobile target-position data |
| US10/729,225 US6983205B2 (en) | 2002-12-06 | 2003-12-05 | Compressing location data of moving objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN02154031.4A CN1278501C (en) | 2002-12-06 | 2002-12-06 | Compressing method and compressing equipment for mobile target-position data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1507171A CN1507171A (en) | 2004-06-23 |
| CN1278501C true CN1278501C (en) | 2006-10-04 |
Family
ID=32778650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN02154031.4A Expired - Lifetime CN1278501C (en) | 2002-12-06 | 2002-12-06 | Compressing method and compressing equipment for mobile target-position data |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US6983205B2 (en) |
| CN (1) | CN1278501C (en) |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005122733A2 (en) * | 2004-06-09 | 2005-12-29 | James Bergin | Systems and methods for management of contact information |
| US7672778B1 (en) * | 2004-07-20 | 2010-03-02 | Navteq North America, Llc | Navigation system with downloaded map data |
| US20070294584A1 (en) * | 2006-04-28 | 2007-12-20 | Microsoft Corporation | Detection and isolation of data items causing computer process crashes |
| WO2009026186A2 (en) * | 2007-08-16 | 2009-02-26 | Cortxt, Inc. | Methods and apparatus for providing location data with variable validity and quality |
| WO2009026556A2 (en) * | 2007-08-23 | 2009-02-26 | Cortxt, Inc. | Location based services information storage and transport |
| US8725407B2 (en) * | 2009-11-09 | 2014-05-13 | United Parcel Service Of America, Inc. | Enhanced location information for points of interest |
| JP5017392B2 (en) * | 2010-02-24 | 2012-09-05 | クラリオン株式会社 | Position estimation apparatus and position estimation method |
| US20120016574A1 (en) * | 2010-07-16 | 2012-01-19 | Research In Motion Limited | Gps trace filtering |
| US20120169535A1 (en) * | 2011-01-05 | 2012-07-05 | Qualcomm Incorporated | Affecting electronic device positioning functions based on measured communication network signal parameters |
| US8929588B2 (en) | 2011-07-22 | 2015-01-06 | Honeywell International Inc. | Object tracking |
| ES2543919T3 (en) * | 2012-03-07 | 2015-08-25 | Telit Automotive Solutions Nv | Contextual data compression for geolocation applications |
| CN103544582B (en) * | 2013-10-14 | 2016-09-28 | 四川品胜品牌管理有限公司 | A kind of method using map denotation logistics distribution scope |
| CN104952248B (en) * | 2015-06-08 | 2017-09-12 | 浙江大学 | A kind of vehicle convergence Forecasting Methodology based on Euclidean space |
| US11023833B2 (en) | 2015-12-30 | 2021-06-01 | International Business Machines Corporation | Pro-active fuel and battery refilling for vehicles |
| US10730626B2 (en) | 2016-04-29 | 2020-08-04 | United Parcel Service Of America, Inc. | Methods of photo matching and photo confirmation for parcel pickup and delivery |
| CN114671025A (en) | 2016-04-29 | 2022-06-28 | 美国联合包裹服务公司 | Method and system for loading, picking up and delivering packages from vehicles by unmanned aerial vehicle |
| CN107133269B (en) * | 2017-04-01 | 2021-06-18 | 中国人民解放军国防科技大学 | Frequent position track generation method and device based on moving target |
| US10775792B2 (en) | 2017-06-13 | 2020-09-15 | United Parcel Service Of America, Inc. | Autonomously delivering items to corresponding delivery locations proximate a delivery route |
| CN108347474B (en) * | 2018-01-22 | 2020-09-29 | 中联重科股份有限公司 | State data compression method and device for mechanical equipment and mechanical equipment |
| US10827310B2 (en) | 2018-09-17 | 2020-11-03 | Phillips Connect Technologies, Inc. | Real-time asset location tracking and monitoring at a low data rate |
| CN111641632A (en) * | 2020-05-28 | 2020-09-08 | 青岛铁木真软件技术有限公司 | Data compression method, system, device and storage medium |
| CN113012224B (en) * | 2021-03-12 | 2022-06-03 | 浙江商汤科技开发有限公司 | Positioning initialization method and related device, equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6374177B1 (en) * | 2000-09-20 | 2002-04-16 | Motorola, Inc. | Method and apparatus for providing navigational services in a wireless communication device |
| US6421602B1 (en) * | 2001-01-03 | 2002-07-16 | Motorola, Inc. | Method of navigation guidance for a distributed communications system having communications nodes |
| US6691028B2 (en) * | 2002-06-07 | 2004-02-10 | Motorola, Inc. | Server-based navigation system and method of operating same |
-
2002
- 2002-12-06 CN CN02154031.4A patent/CN1278501C/en not_active Expired - Lifetime
-
2003
- 2003-12-05 US US10/729,225 patent/US6983205B2/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| US20040158398A1 (en) | 2004-08-12 |
| US6983205B2 (en) | 2006-01-03 |
| CN1507171A (en) | 2004-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1278501C (en) | Compressing method and compressing equipment for mobile target-position data | |
| CN102762957B (en) | Historial traffic data compression | |
| US10715949B2 (en) | Determining timing for determination of applicable geo-fences | |
| CN103794053B (en) | Vague predicting method for city short-distance logistics simple target delivering time | |
| CN110493333B (en) | Method, device and equipment for determining target position point and storage medium | |
| Nibali et al. | Trajic: An effective compression system for trajectory data | |
| CN1149893C (en) | Location determination in telecommunication network | |
| CN110188970B (en) | Shared vehicle management device and shared vehicle management method | |
| CN107357894B (en) | Road traffic facility data acquisition and correction method and system | |
| CN109215338B (en) | Gridding-based Beidou gps vehicle track management system and method thereof | |
| CN1517903A (en) | Map data processing device and centring system | |
| CN104090806A (en) | Method and device for processing upgrade patch and upgrade request | |
| WO2021179642A1 (en) | Position track determination method, device and system, and terminal and storage medium | |
| CN104539750A (en) | IP locating method and device | |
| GB2536561A (en) | Geographical positioning in time | |
| CN101216855A (en) | Message kind business accumulated data caching method and system | |
| CN102692635A (en) | Method for selecting global positioning system (GPS) positioning and base station positioning during store visit | |
| CN101196916A (en) | Method and device for fragment storage file | |
| US7702699B2 (en) | Dynamic data stream histograms for large ranges | |
| US11039273B1 (en) | Method and system for generating geofences using an iterative process | |
| CN107357797A (en) | A kind of information-pushing method and device | |
| CN111641924A (en) | Position data generation method and device and electronic equipment | |
| US20070005921A1 (en) | Segmentation management using a rolling window technique | |
| US8280395B2 (en) | System and method for updating information using limited bandwidth | |
| CN101141758B (en) | Method of managing latitude and longitude of base station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term |
Granted publication date: 20061004 |
|
| CX01 | Expiry of patent term |